DeepChem has an increasingly sophisticated set of tools for materials science users (datasets, featurizers, LCNNs/CGCNNs) but we don’t yet have good documentation/tutorials on how to use DeepChem for materials science applications. Here’s a wishlist of a few things that would be great for us to have as a community
- DeepChem materials science tutorials: Most of our tutorials are implicitly/explicitly focused on drug discovery or chemistry applications. A starter tutorial for materials scientists showing how DeepChem can be a useful addition to the matsci ecosystem would be very useful.
pip install deepchem[mat]: A convenient one-line installation for materials science applications could make DeepChem much more accessible for materials scientists.
- A preprint/position paper demonstrating that DeepChem for materials scientists is ready for primetime usage.
Let me know if you’re interesting in taking on any of these projects! I’ll be working on some of these ideas over the next few months as well